Auditory hair cells release the excitatory neurotransmitter glutamate at their basal pole where they form synapses with afferent fibers. The underlying mechanisms that control and regulate the release of glutamate from hair cells, and how this release elicits action potential spikes in the afferent fibers, are still poorly understood. Part of the reason for this is the small size, fragility, and inaccessibility of adult mammalian inner hair cells and their afferent fibers. We propose here to study fundamental aspects of synaptic transmission from auditory hair cells in the adult bullfrog amphibian papilla. This unique in vitro preparation allows us to routinely access single hair cells and their afferent fibers for high-time-resolution patch-clamp electrophysiology and structure/function studies. We propose to use paired recordings of the hair cell and its connected afferent fiber to study multiquantal glutamate release and simultaneously to measure membrane capacitance changes from the hair cell to assay the exocytosis of synaptic vesicles. We will pursue three Specific Aims: First, we hypothesize that hair cells contain three distinct readily releasable pools of synaptic vesicles that have morphological correlates. Electrophysiological and serial electron microscopy reconstruction studies will be undertaken to determine the size, efficiency of release, Ca-dependence, recruitment and recovery rates, and short-term plasticity of these vesicle pools. Second, we hypothesize that the depletion of a small but fast releasing pool of vesicles accounts for the rapid phase of spike firing adaptation, whereas the second and slower phase of spike adaptation depends on vesicle recruitment from the synaptic ribbon and cytoplasm. Paired recordings will be used to determine vesicle release rates, which will then be compared to afferent fiber spike rates evoked by the same hair-cell stimulus protocol. We will also determine the identity and properties of the ionic currents present at the afferent fiber to better understand how afferent fibers trigger spikes. Finally, we hypothesize that evoked multiquantal EPSC amplitudes become effectively Ca-independent when hair cells are stimulated by sinusoidal-like stimuli that mimic pure tone sounds. We propose that this regime of Ca-independent multiquantal release allows spike synchronization to occur at a given characteristic sound frequency even as stimulus intensity changes. This grant will thus lead to fundamental insights on how the hair cell synapse encodes information about the timing and intensity of sound via the rate, latency, and timing of action potential spikes in the afferent fibers.